ABSTRACT The overarching objective of the proposed research is to evaluate cannabis identity, or the extent to which one views cannabis use as part of their self-concept, as a cognitive and developmental risk factor for adolescent cannabis misuse. Adolescent cannabis misuse is associated with numerous negative consequences and is a growing public health concern given rapidly changing policies surrounding cannabis use in the US. Adolescent cannabis use and misuse may increase in states that legalize recreational cannabis use with initial evidence of this occurring in states (Washington) that have already legalized use. Despite these trends, the cognitive and developmental mechanisms thought to underlie cannabis misuse are poorly understood relative to other substances. Recent theoretical models assert that substance identity, or the extent to which one views substance use as part of their self-concept, is a unique and important cognitive risk factor for substance misuse. Substance-related identities can be measured with explicit (e.g., self-report) and implicit (e.g., reaction time) measures; and both uniquely predict misuse, dependence, and treatment outcomes among alcohol and tobacco users. To date, there are no well-validated measures of cannabis identity, which could represent novel tools for adolescent cannabis misuse screening and potential targets for prevention and intervention strategies. The proposed research aims to address this gap and implements a mixed-methods approach across two phases: Phase 1) 48 adolescents (15-18 years old) in Washington will attend focus groups to discuss the phenomenological experience of having and/or developing a cannabis user or non-user identity. Themes from focus groups will inform stimuli development for two novel Cannabis Identity Implicit Associations Tests (CI-IATs), a type of measure used to assess implicit identification with substance use. 10 additional adolescents will then complete structured interviews to pilot-test the CI-IATs and to rate and discuss CI-IAT stimuli comprehension. Phase 2) 150 adolescents will complete three online assessment batteries, each separated by three months, that include the two CI-IATs, an explicit measure of cannabis identity, and measures of cannabis use, misuse, and cannabis use social norms. To test a dual process model of cannabis identity, the incremental predictive validity of implicit and explicit measures of cannabis identity will be examined in models predicting current and prospective cannabis use and misuse. The incremental predictive validity of cannabis identity measures will also be examined relative to a well-established cognitive risk factor for cannabis misuse (social norms). The research findings will advance theory, potentially improve the prediction of adolescent cannabis misuse, and potentially identify a unique cognitive and developmental risk factor for cannabis misuse that may be targeted among future prevention and intervention efforts.